The Value of Distributed Energy Resources for Heterogeneous Residential Consumers

The presence of behind-the-meter rooftop PV and storage in the residential sector is poised to increase significantly. Here we quantify in detail the value of these technologies to consumers and service providers. We characterize the heterogeneity in household electricity cost savings under time-varying prices due to consumption behavior differences. The top 15% of consumers benefit two to three times as much as the remaining 85%. Different pricing policies do not significantly alter how households fare with respect to one another. We define the value of information as the financial value of improved forecasting capabilities for a household. The typical value of information is 3.5 cents per hour per kWh reduction of standard deviation of forecast error. Coordination services that combine the resources available at all households can reduce costs by an additional 15% to 30% of the original total cost. Surprisingly, on the basis of coordinated action alone, service providers will not encourage adoption beyond 50% within a group. Coordinated information, however, enables the providers to generate additional value with increasing adoption. Distributed energy resources (DERs) are an essential part of modernizing and de-carbonizing the grid [1–3] and pose challenges for the design, management, and operation of the electricity system [4–6]. Dramatic changes are expected as consumers adopt behind-the-meter DERs and become prosumers capable of responding to prices and other signals from grid operators [7, 8]. Technology vendors (e.g. Solar City) and DER resource aggregators (e.g. OhmConnect) will play a significant role in the emerging ecosystem by making DERs accessible to smaller consumers and ensuring that they operate those technologies in a manner aligned with their self-interest and compatible with the needs of the grid as a system. The proliferation of DERs will create opportunities for new business models as well. The coming impact of the adoption of energy storage and rooftop photovoltaic (PV) systems by residential consumers is not well understood [9]. The impact of DERs depends on the adoption rate of technology, the operations strategy, and the financial and policy arrangements for system participants. Consumer behavior and the resulting consumption pattern heterogeneity are critical drivers of DER value and govern the interactions between these dimensions. Residential electricity consumption exhibits significantly more diversity than previously believed [10]. Yet this heterogeneity is seldom accounted for in existing studies, which use data from a small number of residential or commercial consumers to evaluate the consumer-side economics of PV [11–15], storage [16,17], or both combined [18–24]. In this paper we propose a simple and scalable methodology to estimate DER value and impact that incorporates consumption heterogeneity. We apply our methodology to provide a first of its 1 ar X iv :1 70 9. 08 14 0v 2 [ cs .S Y ] 2 5 O ct 2 01 7 kind assessment of the value of these technologies to residential consumers, technology vendors, and aggregators under various business models and policy arrangements. We focus on PV and batteries because they are commercially available and gaining traction as their costs decrease [25–29]. Our study accounts for adoption rates and consumption heterogeneity in unique ways to identify tipping points for impact. We utilize a large dataset of hourly power consumption recordings for residential consumers in Northern California to capture heterogeneity. Based on this data we build various models to assess the value of storage and PV to consumers in the form of bill savings. We estimate the value that can be delivered by entities that provide improved information to and enable resource sharing among residential consumers. Households rely on information about future consumption and generation when deciding how to operate their DERs. We model how constraints on this information impact the value of DERs to households. This analysis gives an estimate of the value of services that improve the accuracy of information available to the households. Communication technology enables aggregators to coordinate and share DERs among a group of households. We analyze the value that these coordination services can generate, which depends on the pattern of technology adoption by households and the sharing mechanism. 1 Assessing value from data Storage devices enable households to shift their energy usage in time, and rooftop PV generates electricity that households can consume directly or sell back to the grid. Both technologies allow households to reduce their electricity costs. We estimate how much households and groups of households could save on their bills if they adopted a 5 kW rooftop PV system and a 7 kWh storage device1 and operated these devices to minimize their electricity costs. We take a household’s hourly smart meter data as its inflexible end-use consumption. We provide an a-priori snapshot of potential bill savings. Thus, electricity prices are exogenous in our model. Similarly, we assume that households do not significantly alter their electricity consumption behavior when they adopt DERs or when they face differing rates. Refer to the Methods section for detailed explanations of the data sources and analyses. 2 Value of technology for households We define absolute savings for a household, Sa,i, as how much its electricity bill decreases when it adopts PV and storage. A household’s normalized savings is Sn,i = Sa,i/(1 Li), its absolute annual savings divided by its original total energy consumption. This gives a cents per kilowatt-hour savings estimate for each household. Relative savings is the ratio between a household’s normalized savings and that of the top household under a given policy, i.e. Sn,i/ ( max i Sn,i ) . Relative savings are always between 0 and 1, and they enable a comparison of the concentration of benefits under the different pricing policies. We compute savings under three pricing policies, incorporating time-of-use (TOU), wholesale, and dynamic rates, as listed in the table in Figure 1. We conducted five sensitivity studies with varying PV and storage device sizes. Most of the qualitative results of our study were unchanged.